GCMS Is it necessary to separate peaks ?

[seamoro]

I am bit a bit confused...somebody told me it wasnt necessary to separate peaks with a GCMS! that the MS will scan for the particular ions of compound. So it is not necessary to separate the compounds to accuratley quantify and identify....this is false....right??

[Consumer Products Guy]

Separation of peaks is a great thing. However, when two components co-elute or have poor separation, there can be ways to still quantitate each. For example, if Component 1 had a nice peak ion at 220 m/e and Component 2 had a nice ion at 157 m/e, and 1 was clean at 157 m/e, and 2 was clean at 220 m/e, then one could extract the 220 ion and quantitate Component 1 from that, and then extract the 157 ion and quantitate Component 2 from that.

Similarly, SIM mode could be used to do similar by only looking at speciied ions.

We had a GCMS assay for panthenol that used extracted ion as the panthenol coeluted with a matruix component.

[Don_Hilton]

You like to have sufficient separation that you avoid chromatographic effects that make peaks to be poorly shaped or result in ion supression. Sometimes if you have a coelution with a compund that is present at a sufficient quantity that it will change the nature of the stationary phase a bit - changing chromatographic properties across the peak for your analyte of interest - and the peak may be defocused on the column in such a way that it has an odd shape. If you have coelution with a great quantity of a compoound that most of the ionization is on the coeluting material and the peak (or a portion of the peak) for the analyte of interest is diminished in intensity (this is ion supression).

As long as you avoid these two issues, you can have a very good method quantifying specific ions from overlapping peaks. Many standard methods work this way.

[Peter Apps]

As long as the retention times and spectra are not exactly the same you can get a long way with deconvolution of GC-MS data. I use AnalyzerPro from Spectral Works on very messy data and it is very good at disentangling overlapped peaks. You can get a free trial.

Peter

[James_Ball]

For analysis like EPA 8260 and 8270 where your analyte list can be over 100 compounds and you are running them in less than 30 minutes you will have many coelutions but can quantitate them using specific ions. This is a very common practice in environmental analysis. When you are looking for trace compounds and you have a hydrocarbon matrix, like when looking for PAH compounds in oil, you will also run into this type of problem, and the deconvolution software comes in very handy. Pesticides in plant tissues is another example of the power of MS to extract useful data from a mass of coeluting peaks.

Separation is always preferred, but not necessary as long as you have a clear quantitation ion to choose.

[MS-Omics]

We apply the PARAFAC2 model to deconvolute GC-MS data, which we find very powerful.

PARAFAC2 highlights:
-Simultaneous processing of all GC-MS runs from an experiment
-More compounds → more value
-Clean MS-Spectra → better NIST library hit
-Less noise in final data → more conclusive results

One our homepage you can find an example of a graphical user interphase that we have developed for analyzing GC-MS data with PARAFAC2. We offer both data processing service and a training course as part of our services.

Best regards,
Morten

[MSCHemist]

In general yes, It is quite common for compounds to have very similar EI fragment ions as many compounds are carbon hydrogen and generate the same fragments. Also compounds generate a lot of fragments so even if one compound has a 71 and the other a 72 isotopes will cause them to have some of the same ion. Indeed in many cases when two compounds coelute on the GC, you can pick unique ions to quantitate each one cleanly. Basically the MS ion seperation gives you some wiggle room with resolution but should not be counted on if it is possible to get chromatographic separation.

[thohry]

For quite simple samples with few components, it's ok not to separate completely chromatographically. But for complicated sample, the separation is important.
May be with GC/MS/MS, the chromatographic separation is not so important. .

[mckrause]

If you're looking for isomers you have to separate. If you're looking for disparate analytes then you don't have to separate. We use RGA systems (residual gas analyzers) all the time; they have absolutely no separation, but give us beautiful O2/N2/CO2/Ar data, which is all we need from them. However, if you're trying to determine, say, benzo(b)lfluoranthene and benzo(k)fluoranthene then you'd better be able to get them separated!